When you bring in a design agency you enable them to become an extension of your team, to understand the kind of person, group or company you are, the tenor of your interactions, and to experience things that can’t be placed in a requirements document. In the 80’s, Apple Computer used companies like Frog Design and Fitch to design the look and feel of their hardware up until the 1992 when Apple pulled their design in-house under Jonathan Ive. While the Apple II and Macintosh computers showed good and innovative design, the work that has happened since blows it away both in terms of impact to consumer spending and critical acclaim. So how do you achieve that kind of success?

Step 1: Empower your designer

So you have a new graphic/interaction designer, or design firm starting up, and you want to get them going as fast as possible. What do you do? Do NOT send over a requirements document first. Instead, start by having them understand your objectives – first hand. Get them charged up about the project and tell them how you got to this place, what the history was that led you to start up this project. Be honest about past mistakes, and about future desires. Don’t be afraid to say that you want to be “more Apple like” – it’s ok to want to be good. Designers, like actors or sales people, are more sensitive than the average person, and to get the best from them you need to make sure they’re in a place, emotionally, where they can succeed. It doesn’t mean you need to treat them with kid-gloves, but you can’t be disparaging or simply have neutral reactions. If you have a hard time smiling, then directing designers is not for you.

Step 2: Get inside their head

Next, ask them questions so that you understand how they think, what they care about, what elements they focus on, and how they produce designs. In this process not only will you get an understanding about the kinds of things they care about but also understand how their approach aligns (or doesn’t) to your goals. You may discover that one designer wants to dive right in, while another wants to talk about abstract ideas before putting pen to paper. Learning about your designer will better help you communicate. After this introduction period, you can present the requirements doc, and make sure they understand the end-state of the final application – even if they’re not creating the whole thing! Make sure that each designer can directly connect how their contribution connects to the success of the project.

Step 3: Don’t think that design time is linear: 50% is not half way done

Just because the schedule says you should be half way done since you’ve burned half the time, don’t assume that your designer will be half way done with the work. Design is a hockey stick curve: slow at first, seemingly slow at the end, and then a swirl of good activity. And you want it that way.

Here’s the problem: if you get a designer to produce something too quickly that appears to be too finished, you start to set the designers mind in a particular path which can be hard to break. Instead, allow them to come up with ideas, fragments of ideas, and sketches. If you’re not used to working with designers this process may seem to take too long.

Step 4: Keep them on schedule

Keep the designers on schedule by ensuring that they’re setting their production deadlines, and review cycles. They have to understand the larger time constraints, of course, but by letting them dictate the progress you’ll set up a reciprocal relationship and they will produce their best work. Note: This is very hard to do with a junior designer and only the very best young designers can transcend this, which is why it’s often easier to work with someone that has had agency experience.

Step 5: Make sure they tell you when you have a stupid idea

Good designers push back. They may be wrong. You may be wrong. But the discussion will yield better results. Here’s why: A designer who pushes back, but who is wrong (for example, they don’t want to show a widget because it’s ugly, but you think it needs to be there because you’re producing a demo to show an investor) will learn more about why you want certain elements to be visually prominent. When this conversation is resolved the designer will then return to the project and align all the other elements that could help achieve your goal – either pushing other widgets into focus – or making the ugly object seem less ugly and feel integrated into the whole.

Step 6: Be available for reviews – even at odd times

Designers don’t ever stop working. They may not be at the office but they’re still thinking about your problem and getting inspired at all times of day. When they have a breakthrough idea, or they’re struggling through designers-block, they need to get feedback or inspiration quickly. Make yourself available to your designers – even just for a quick phone call, and you’ll see how those moments translate into significantly improved end results.

Step 7: Get more than one designer

Designers generally like to collaborate with other designers, which is why it’s best to have 2 designers who are available to each other. They don’t have to have to be from the same discipline – so get a human-factors designer, and a graphic designer, or an interaction designer and a information-architect together – and even if they’re working on separate projects they’ll be able to help fuel and extend one another’s thinking.

Step 8: Push your designer to solve the unsolvable

Designers exist to push the limits of what’s been done. It is their obligation to determine how to satisfy two seemingly irreconcilable goals. If an executive says “Make the text bigger and smaller” it might (reasonably) sound like an unsolvable problem. But if a designer thinks about the reasoning behind the comment, they might figure out that the reviewing exec means: make it bigger (so I can read it more easily) but make it (seem) smaller (in relation to other elements of the page so it doesn’t seem as prominent). Designers face these challenges all the time and they thrive on solving those problems – so push them to solve them, and help them solve them by looking for a context that rationalizes both demands.

Step 9: Don’t ask for a banana when you want a duck

The best way to kill a designer’s energy is to send them off to solve a problem, have them produce something you really like, and then force them make changes which, in totality, cause them to lose the core of their design. So please don’t say “I love it” (which you do) and then say “but can we make the text a different font, the colors darker, remove the center column, put the logo on the upper right instead, and make my name more prominent?” Instead, if you like what they did, but it’s not working on a number of levels, tell them what you liked about it, then ask for another “treatment to the design” that satisfies your additional objectives. See Step 8.

Step 10: Thank your designer

Wrapping up a project the right way is critical to fostering a good relationship with a designer. Designers need closure, and a post mortem will allow you to give constructive feedback to the designer as well as provide an official way to give positive accolades from others who

In real (our non-web) life, we love it when people expertly craft objects, and then allow us to determine how they look when we use them. While Chris and I might buy the same tie, Chris might tie a classic full-Windsor knot, while I make a Kent knot. Joyce prefers jeans and sneakers, while Alyssa wears a flowing dress with flip-flops. Each person gets to choose from the same set of objects and select which ones are appropriate to use for any particular situation (as well as determine how to wear them). On our websites, as in life, we want the right kind of flexibility that doesn’t violate our brand but also allows us to serve each person as if they were a unique population unto themselves. A personalized, styled experience. So how can we do it?

A Flexible Architecture for Building a Rule-Driven Modular UI

If you use a flexible architecture to connect the data and the user, you can create a system that gives you the best of both worlds: a user interface comprised of a structured set of modules driven by rules, in which each user gets an experience appropriate for them. Of course, you need to have a system that can synthesize the big 3:

1. Record data - like product or part data
2. User input - what actions the user is taking at that moment
3. Business rules - what actions or data you wish to promote

…and be able to create the right UI on the fly.

That part is well understood. If you’re using Endeca you have the mechanism and the controls to do just that, but the hard part is putting it all together in a rich and robust way. So, here’s a framework for an easy way to allow you to design, and implement a system that allows your designers to think flexibly while allowing the business user to control multiple segments in very specific ways, all driven by rules. The only change in the presentation layer is to build it modularly. In a site that has a user interface which is modular and completely rule-driven, you can do lots of interesting things just by setting up rules.

A bunch of sample rules based on a users action/context

Here’s a variegated set of rules that reflect just the tip of the iceberg you could create:

• If a user navigates to a hot new product, the theme of the page could change to reflect the point brand of the product line.

• If the user tends to use a tag-cloud a lot, the tag-cloud widget can move to a central location for just that user.

• Does that user appear to be navigating through content designed for an expert? Then technical documents could appear in a spotlight area and push away marketing literature.

• Does the user seem to be in a category of people who wouldn’t click on half of the tabs in the mast head? Then employ a modified mast-head that’s more targeted to that particular user segment.

• Does the user appear to be pogo-sticking between 2 product records? Maybe they don’t see the “compare” check boxes – so present an alternate treatment automatically.

• Would different user populations benefit from different video-presentation widgets? Then use a rule that chooses the best one based on the context (maybe one for showing just a couple of videos, one that can hold lots of videos and links, another with a vertical orientation, etc.)

And because it’s all modular, and rules based, you can now enable millions of user-interfaces that all happen automatically, for each user, in a specific and targeted way.

Social networking sites and popular Internet portals have lead the way in modular site design. Using industry leading portal systems developers encode functionality into widgets such asSharepoint webparts or Liferay portlets, which allow the layout to be dictated by the user. The user can move the widgets around, choose which widgets to add or delete, add severaldifferent widgets of a similar type or even choose a theme that controls the look of the page. Now imagine that the look, layout, the data that populates the widgets are all controlled by your business rules and done in a way that targets each user individually… that means the system is creating a unique user experience for each person. The hard part of making this is that you have to be able to structure a document that allows you to capture the desired behaviors of each module, in which each module is triggered by using all kinds of data, like user profiles, user actions, etc. To make it easy, we’ve put together a simple structure that you can use as a jumping off point.

Here’s an easy way to deconstruct the problem

1. Determine who your users are, and break them up by segments (such as, novice cooks, expert chefs, foodies)

2. Determine the major areas of the site (recipes, cooking equipment, cooking theory)

3. Determine the major actions that users would do (look through a series of recipes, compare cooking products, read restaurant reviews)

4. Construct templates that lay out a structure that will form the basis of all the pages (template for looking at lots of video content, template for comparing products, template for a landing page designed for a new or less active user of the site, template for a landing page designed for an active user with a smaller splash image)

5. Determine which widgets will live in each template, and how those widgets looks and work (spotlight widget that shows images, spotlight widget that shows documents, video widget that plays a video and has thumbnails of related videos)

6. Determine the behavior of each widget in context.

How to write up a simple document

Lay out these concepts in a grid on a spreadsheet, and you’ll see how you can quickly you can start to fill it in. Basically you’ll have three columns for the input actions that set the state of the application as to who is where and what they’re doing (numbers 1, 2 and 3) and three columns for the output actions that describe which template(s) to use, which widgets to employ in the template and what the behavior of each widget should be (4, 5 and 6). What you’re building is a set of rules that determine what’s shown, when it’s shown, what content goes in each widget and what controls that content. Sounds complex? It’s not, just start with something easy.

Starting off easy

You can always start with a rule that simply covers all cases – so perhaps you start by employing a rule for “Everyone”, that occurs “All The Time”, and uses the “Normal Corporate Header Template”, which includes an image called: “Company_Logo” and set of tabs for the major areas of the site. That’s an easy rule, and would produce a standard functioning header. The nice thing about using a rule that controls how the header fires is that you can change it on a whim. Perhaps one day you want the header to look different (winter time? Use an image of the logo with snow on it) or to function in a more personal way (add a “Things that <user name> might like” tab) or change it based on where the user came from (add “Welcome <company name>, we’ve applied your discounts to all the prices you see.”)

And of course, a site that’s easier to modify, based on rules and templates makes it easier to ensure consistency and enforce a style across every part of your website.

-David Pogue, tech reviewer, on his omission of reviewing a mini camcorder called the Flip, costing $150 that now has 13% market share.

What is it in certain people that make them believe greater complexity somehow equals more power? Or think that products which are feature rich are more appealing than ones which have been carefully edited? Or there’s low value to be had in telling a clear and cogent story that deliberately trades off covering all the elements of an idea in favor of a story heavily edited but emotionally compelling and easy-to-understand?

To believe that more is more, is the fatal flaw of the techno-focused (but not tech-savvy) designer. How does it happen that some people are so misguided? It comes down to embarrassment. People are embarrassed to believe that if they aren’t struggling to create more stuff, they offer the world less value. Is a term paper better because it’s longer? No. Is a store better because it carries more items? No. And, (do I really need to say this?) is a software application better because it has more features? Of course not.

Embarrassment is why a well-respected reviewer misses a very hot and successful product, only to extol its virtues in his (if-he-wrote-it-any-later-it’d-be-a-retrospective) review. The reason Pogue didn’t review it? He cites that “…the Flip looks like a cheesy toy that no self-respecting geek would fool with, let alone a technology columnist.” He was too embarrassed to believe that the simplicity that this new tool offered would be engaging to the target demographic.

And just last week, one of my more tech-centric graduate students said, when joining a group discussion about the Apple brand, “Computers aren’t toys, and shouldn’t be made like one.” Really, where do I begin? Toys are bad? Do we not want to be surrounded by toys every day at our tech jobs? Sounds like he’s embarrassed to admit that having fun can be serious work. What these people don’t understand is that achieving the right kind of simplicity is remarkably difficult. You don’t just cut out features, or go lo-fi, or paint it glossy white.

With the success of products that range from the Polaroid Instamatic camera, TiVo, and almost everything Apple produces, it’s clear that good design is one of the most powerful ways for any company to differentiate from its competitors. But don’t think that those companies only focus on design as it relates to their products: the importance of design is pervasive throughout successful corporations. As expected, the Apple campus in Cupertino is significantly easier to navigate than the Microsoft campus in Redmond.

This week, Robert H. Scott, chairman of the board of directors of the DeCordova museum in Lincoln MA, was quoted in the Boston Globe saying “there’s a developing theory in the management that people who are experts in design can add a great deal in the running of organizations. Many young people who used to go to management schools are now going to design schools.” Why? Because good designers solve complex business problems, manifested in all kinds of ways; spoken word, and written text, visual and auditory design, and even event planning – all accomplished by making the complex, seem simple. The hard to understand, “grok-able.” The dry and dull, sexy and desired.

So why do so few companies strive for good design? It’s because achieving elegant simplicity is astoundingly hard. And sadly, a bit embarrassing for some.

Dealer: [you should have us do] the “A” service, it’s a routine service that includes an oil and filter change. We will also check computer systems for fault codes. $289.00 for this service.

Me: So its $280 for an oil and filter change?

Dealer: It actually includes a safety check with a computer systems check. Checking SRS, ESP, Fuel management system, alarm system and lighting system faults.

Me: Wow, doesn’t sound too compelling from a cost perspective. Is there something I’m missing? And how much is an oil and filter change usually?

Dealer: The oil and filter change is $99.95. Service should be done each year. This is the best check if your vehicle is going out of warranty soon, so that’s what I would recommend to you.

Me: ok - can you just help me understand what happens for the extra $180 of the A service? (you described it so easily that it makes it sound like someone simply hooks up the computer and checks fault codes… what else do they do?)

Dealer: This is just what the vehicle is calling for. You don’t have to perform this service if you desire not to. I was just stating what your vehicle is asking for. Please let me know if you would like an oil and filter change or the full A service.

The problem here is that I have some questions that he needs to answer, and since he can’t articulate the answers, I wasn’t willing to buy into the more expensive service.

The dealer hadn’t successfully convinced me of his reasoning so we never got alignment. As a result, in this case, the project never started. But it could have been worse: I could have agreed to the expensive service and been surprised by the deliverable. This issue is no different in the work environment when making small or large changes to a system. You have to be able to articulate the rationale behind the requirements so that your designer can be aligned with the objectives of the project and contribute maximum value.

Here’s a short list of questions that a designer (and, in fact, every person on a team) needs to be able to answer to ensure things are kicked off on the right foot:

About the Goals

1. What’s the primary goal of this app/modification?
2. How will the app support the brand and the company?
3. For an Endeca solution: which elements of Endeca’s technology, and other related technologies, are suited for meeting the goals?
4. What makes this solution (or modification) unique and how should that be showcased (or should it)

About the Users

5. Who are the target users?
6.What are the core activities and goals of the various user groups, and what currently makes them successful in their job?
7. What other tools/manual operations do they do now (or perhaps, will they have to keep doing) that exist outside the new system/modification?
8. What are the most important features for your users?
9. What’s the environment where the system will be used? Physical (home, in-store kiosk), or emotional environment (relaxed user, frustrated user, in a hurry).
10. What’s their familiarity with the market/terminology/similar tools?
11. How do you think different users will use the app? Will some read lots of reviews, others compare lots of records, while others perform analytics and drill down through charts?

About the Outside World

12. What other applications exist that are similar to this one? How are competitors handling this same problem?
13. What features of those applications are the most well-liked? least-liked

About Achieving Success

14. How will success be measured? (Quantitative as well as qualitative)
15. Who are the internal stakeholders who need to be kept in the loop for sneak previews and approvals?
16. What are the foreseeable challenges and constraints in this project?
17. Has your company tackled similar challenges in the past and if so, how were they overcome? Can you leverage those solutions/lessons?
18. Who are the points of contact for questions/clarification?
19. What factors are motivating the timeline?